Effective action in spherical domains (Q1328229)

In an earlier work [\textit{P. Chang} and the author, Nucl. Phys. B 395, No. 1-2, 407-432 (1993)] we have shown that the \(\zeta\)-function, \(\zeta_ \Gamma(s)\) on orbifold-factored spheres, \(S^ d/\Gamma\), for a conformally coupled scalar field, is given by a Barnes \(\zeta\)-function, \(\zeta_ d(s,a\mid {\mathbf d})\), where the \(d_ i\) are the degrees associated with the tiling group \(\Gamma\). The free-field Casimir energy on the space-time \(\mathbb{R} \times S^ d/\Gamma\) was given as the value of the \(\zeta\)-function at a negative integer which evaluated to a generalised Bernoulli function. In the present work we wish to consider the effective action on orbifolds \(S^ d/\Gamma\) which this time are to be looked upon as Euclidean space-times. In particular we will discuss \(d = 2\) and \(d = 3\), concentrating on the former results. The simplifying assumption in our previous work was that of conformal coupling on \(\mathbb{R} \times S^ d/\Gamma\). This made the relevant eigenvalues perfect squares and allowed us to use known generating functions to incorporate the degeneracies. From the point of view of field theories on the space-times \(S^ d/\Gamma\), retaining this assumption would be rather artificial. A more appropriate choice would be minimal coupling, or possibly conformal coupling, on \(S^ d/\Gamma\). (These coincide for \(d = 2\).) The quantities in which we are interested are \(\zeta_ \Gamma'(0)\) and \(\zeta_ \Gamma(0)\). The latter determines the divergence in the effective action and the former is, up to a factor and a finite addition, the renormalised effective action (i.e. half the logarithm of the functional determinant).